Generative Sountracks for Oral Storytelling

Evoking emotions is one of the main goals of artists when they are creating an artwork. Thus, it is fundamental for artists to be able to analyze and manipulate art pieces in order to express desired emotions. This makes it important for AI-based creative agents to have or mimic the same abilities. This project explores the problem of soundtrack generation for oral storytelling as a method for manipulation of artistic pieces in order to evoke specific emotions in humans.

Procedural Level Generation for Physics-based Puzzle Games

Physics-based puzzle games add an extra layer of difficulty to procedural content generation (PCG) because their mechanics are based on "realistic" physics. Thus, evaluating feasibility and quality is harder and typically requires simulations. This project consists of designing PCG methods capable of generating feasible and interesting levels for physics-based puzzle games, such as Angry Birds. To support the experiments in this project, it was implemented an Angry Birds clone, which was coded using the Unity engine and the original art assets. The clone source code is available for download via GitHub.